Intimate partner violence is widespread in the United States and causes serious physical and psychological damage to the victims. Although much of partner violence is reciprocal, women suffer far more damage from victimization than men. Alcohol is implicated in violent relationships, both as a chronic risk factor of and an immediate threat during violent episodes. Little is known about the daily interactional processes between partners, the immediate role of alcohol use in those processes, or the context within which violence and drinking occurs. Our program of study uses nonlinear dynamics analysis to look inside the black box of violent interchanges within couples. During patient waiting times in a primary care clinic, we propose to enroll 20 couples with a recent history of intimate partner violence and ask them to respond to a daily telephone survey for 60 days. Participants will also complete baseline and end-of-study surveys addressing risk factors and outcomes of victimization and alcohol abuse. Daily, they will report on partner violence, alcohol use, arguments, hassles, stress, and marital closeness. Nonlinear systems analyses will determine the patterns of violence and alcohol use and the influence of other daily factors. Several measures to promote safety will be in place. The specific aims of this exploratory study are to: 1. To explore the ethical feasibility of conducting a large-scale longitudinal study of daily intimte partner violence and alcohol use, and their predictors. We will: estimate recruitment, retention and missing data rates; estimate effect size to aid in sample size determination; and evaluate the impact of study participation on subjects, including safety. 2. To conduct preliminary testing of 20 couples in violent relationships in order to: A. assess the concordance of partners' reports of violent behaviors; B. estimate linearity/nonlinearity in daily violent events and alcohol intak over 60 days; C. understand the covariability and stability of violent events between male and female partners and between violence and alcohol intake; and D. prepare for system dynamics modeling simulation by identifying feedback mechanisms that could explain the dynamic patterns of husband-perpetrated violence, treatment-seeking, and marital separation. This line of inquiry seeks to unravel the complex reciprocal effects of husband-wife alcohol use covariability, violence covariability, and the proximal alcohol use-violence relationship. Ultimately, these findings could have significant clinical applications, suggesting interventions based on couple dynamics or leverage points for potential change. Identification of specific patterns of violent events may permit self- monitoring for warning signs by at-risk women, risk screening and brief interventions by health care professionals, and/or recognition of optimal times to intervene by victim advocates.